Here's an example pipeline that computes the 52-week high for each asset (using end-of-day daily prices). I also included a version of your factor that uses yesterday's close price as the 'current price' in your description.

Let me know if this helps.

thank you very much this is exactly what I was hoping to achieve. I was trying to construct my own custom factor initially but failed at first attempt. I have a question regarding the 'close_prices' parameter you use. what exactly is that and in general what is the values parameter supposed to do [def compute(self, today, asset_ids, out, values):]

also how comes we can initiate the window_length right at the beginning and not when we call the custom factor in our pipeline.

is there a video or any other resource that could guide me better on this other than the documentation. Thank you for helping

The close price is coming from USEquityPricing.close.latest which is a column (close) on a pipeline dataset (USEquityPricing). Each day that a pipeline is computed (in the morning), it will retrieve the close price from the end of the previous trading day.

If you haven't read through it yet, the Pipeline tutorial is the best resource for learning about this stuff. Lesson 10 focuses on CustomFactors and explains the values variable(s), but I'd recommend starting from the first lesson to understand the basics of pipeline first.

Let me know if this helps.

Just cause I'm finding this, the links to the pipeline tutorial are incorrect. Here's the corrected link to lesson 10 https://www.quantopian.com/tutorials/pipeline#lesson10

Aren't you getting problems with USEquityPricing.close.latest during splits? I see it not being split adjusted (it jumps). Or does using it a factor+a window take care of that?